基于惯性传感器的老年人姿态监测装置设计

为解决人口老龄化社会所带来的老人跌倒事件频发,减轻跌倒等异常状态对老年人的伤害,研发了一种基于惯性传感器的穿戴式姿态监测装置。该装置基于姿态解算特性,通过采集三轴加速度计、三轴陀螺仪和三轴磁力计的数据,以多种运动特征为依据进行姿态判别,不仅可以检测跌倒状态,还可对老年人的日常姿态及位置信息进行连续监测并进行步数统计,利用GPRS网络向服务器上报数据。装置中加入了语音推送功能,可对服务器端推送的天气信息、问候信息等进行语音播报。试验结果表明,该装置可对佩戴者的姿态及位置信息进行连续监测,当出现跌倒等异常状态时,可实现实时报警及对事故发生地的准确定位,姿态识别的准确率不低于98%。     

基于三维加速度传感器设计的跌倒检测

背景:老年人的机体会随着年龄的增长而发生一定的变化,急、慢性疾病的病理性改变可能会影响感觉输入、中枢神经系统功能和骨骼肌肉力量的协调,从而导致老年人发生意外跌倒的事故也越来越多。如果在日常生活中,老年人发生跌倒的时候,能够及时发现,就能避免更严重的后果。目的:设计一个基于三维加速度传感器的跌倒检测装置。方法:结合ZigBee无线通信技术,设计一个基于三维加速度传感器的跌倒检测装置。将采集的加速度信号通过ZigBee方式无线发送至上位机进行数据分析处理。结果与结论:实验证实,该装置可以实现对跌倒发生的准确判断,并能对发生跌倒的类型进行分类,使得老年人在发生跌倒后可以及时得到医疗救护,并为医生对老年人的健康状况的判断提供了可参考的信息。     

基于多传感器数据融合的跌倒检测算法

目的跌倒在老年人生活中是一种常见的现象,是致使老年人发病和死亡的主要原因之一。实时的跌倒检测系统能够及时报警,缩短等待救治的时间,减少由跌倒引起的意外伤害。可是,在大多数的跌倒检测系统中,人们仅利用加速度计设计检测系统,基于单一数据的算法不能完整表征跌倒时身体姿态变化的信息。为此本文拟采用陀螺仪和加速度计的数据设计跌倒检测的算法。方法首先介绍了利用MEMS惯性传感器设计置于腰间的可穿戴的跌倒检测系统,然后对跌倒的规律进行了分析,基于此提出了基于多传感器数据融合的跌倒检测算法,即通过数据融合的技术提取出身体加速度及其动态量和静态量、加速度变化量、身体姿态角、角速度绝对值之和等特征参数,利用多参数设计了基于阈值判定的跌倒检测算法。结果收集10名志愿者做模拟跌倒以及日常活动的数据,对算法的有效性进行验证,取得96.67%的灵敏度和97%的特异性,并且此指标高于Kagans等算法的结果。结论本文提出的算法在跌倒检测中具有较好的有效性及优点。     

基于MSP430及安卓系统的老年人生理信息监护系统

目的：随着老龄化的不断加剧,老年人的监护需求量也在不断的增加,为了解决相应的老年人监护问题,介绍了一种基于MSP430单片机和安卓系统结合的老年人生理信息采集监护系统的工作原理。方法：以MSP430为核心,配以血氧、体温等传感器模块对老年人生理信息进行采集,利用传感器技术与嵌入式技术相结合。设计了生理信息采集终端。设计中采用了跌倒检测,在老年人出现跌倒时,能够及时通知监护人员,防止出现救治不及时的情况,同时还利用蓝牙通信,将采集的信息上传到基于Android系统的智能手机终端。利用Android系统上传数据到服务器,为老年人建立生理监护系统。结果：对设计进行了实验测试．经试验表明系统能够正常的进行数据采集,体温测量数据误差在±0．3℃,血氧饱和度的测量在正常范围95％-98％之间,跌倒检测很准确的检测到了跌倒状况,满足测试的标准,达到了设计的预期效果。结论：该系统能够很好的实现老年人的生理信息监护．对老年人的血氧、体温进行了很好的监测．在老年人出现跌倒时能够及时的进行报警,得到救治,具有良好的扩展性和广泛的应用前景。     

基于极限学习机的跌倒检测分类识别研究

由于年龄和身体条件的限制,在老年人群中跌倒是非常普遍的现象。因此,根据老年人跌倒的运动特征,远程监测他们在各个时间段的状态,以便在其摔倒或突发状况时及时采取措施显得尤为重要。针对人体运动状态进行监测,分析人体运动学特征,提出基于极限学习机的跌倒检测算法。运用三维加速度传感器采集人体的三维加速度值,建立跌倒检测特征模型。在此基础上,建立基于极限学习机的跌倒检测分类器,完成对老年人的计算机辅助跌倒检测。实验数据共540例样本,选用了不同数量的训练集和测试集,其中440例作为训练数据,其余100例为测试数据。测试结果表明,准确率为93%,敏感度为87.5%,特异性为91.7%,具有良好的分类性能。在对分类训练的运行时间方面,基于极限学习机的跌倒检测方法与传统的机器学习方法相比具有明显优势。     

一种老年人移动健康监护系统的研究

目的:设计一种老年人移动监护系统,实现移动心率检测,跌倒检测,紧急情况自动电话呼叫与GPS定位,减少突发事件对老年人的健康威胁。方法:采用分层的体系结构,第一层为传感器层,用于采集并实时分析ECG信号以及加速度数据,第二层为手持终端,用于汇集各传感器数据,综合判断并实现数据远程传输、呼叫以及定位,第三层位于远程服务器,实现数据库管理,数据显示等。结果:分层的体系结构有利于系统功能的扩展,有利于传感器单元的可穿戴性。结论:基于无线传感器网络与GPRS网络的移动健康监护系统能够有效实时地对老年人的健康与安全状况进行监护,对老年人生命安全保障具有重要意义。     

基于三轴加速度传感器的频率自适应心脏起搏器运动-心率实验数据采集分析

本研究基于三轴加速度传感器系统和心率采集系统,通过征集98名志愿者,进行269次Bruce-Q运动实验,同步采集志愿者运动时的X(左右)、Y(上下)、Z(前后)三轴向加速度数据和心率值,建立三轴加速度运动—心率数据库;分析实验结果可知,人体各个轴向运动加速度与心率线性相关,运动后心率恢复具有呈指数下降的规律性;结果表明,在运动时Y(上下)轴、Z(前后)轴比X(左右)轴更能反应人体真实运动状态。     

基于隐马尔可夫模型的老年人跌倒行为检测方法研究

随着社会老龄化程度的加剧,老年人的安全健康监护需求日益增加。跌倒行为在老年人日常生活中比较常见,它会给老年人带来严重的身体及心理伤害。因此,跌倒检测对于保护老年人的健康及安全具有重要意义。针对跌倒的运动过程,分析人体加速度变化特征,提出基于隐马尔可夫模型(HMM)的跌倒检测方法。将人体跌倒的加速度信号提取为加速度观测序列,并以此为训练样本训练隐马尔可夫模型,建立跌倒过程的概率模型进行跌倒检测。在验证实验中,采集10名志愿者共300例样本,采用5折交叉检验方法,对模型的有效性进行验证。验证结果表明,该方法检测跌倒的准确率为98.2%,灵敏度为91.3%,特异性为99.6%,具有良好的检测效果,可实现对跌倒行为的准确检测。     

穿戴式心电及心振信号监测系统的研制

目的 心脏的电活动及机械活动对于研究心脏的生理功能和病理改变具有重要作用,为此本研究研制了一套心电(electrocardiogram,ECG)及心振(seismocardiogram,SCG)信号同步监测系统。方法 采用单导联方式进行心电检测,通过测量胸廓皮肤表面的三轴线加速度和三轴角速度来获取心振信号,以ARM Cotex-M3内核的低功耗微处理器STM32作为主控芯片将数据存储至SD卡,或通过低功耗蓝牙将采集的数据发送至手机端上位机并进行数据处理及波形显示。结果 对人体进行测量以获取ECG及SCG信号,结果表明系统能够实时采集、传输、显示具有较高信噪比的心电和心振信号,并能实现信号的长期离线记录。结论 该系统具有便携、操作简单的特点,能够长期稳定工作,为下一步分析心脏的电活动与机械活动之间的耦联关系奠定了基础。     

跌倒检测中人体头部和腰部相关性的研究

本研究采用惯性传感器对跌倒过程中人体头部和腰部的运动学数据进行相关性分析,由志愿者模拟五种日常活动动作和四种跌倒动作,进行大量模拟试验。发现在实验条件下,人体头部和腰部的合加速度相关性极低或不相关,头部和腰部的合加速度偏离竖直方向的角度在大部分日常活动中相关性极低或不相关,但在跌倒过程中,具有较高相关性。该结论可为跌倒的预测及检测提供一种新方法。     

AngleNav: MEMS Tracker to Facilitate CT-Guided Puncture

As a low-cost needle navigation system, AngleNav may be used to improve the accuracy, speed, and ease of CT-guided needle punctures. The AngleNav hardware includes a wireless device with a microelectromechanical (MEMS) tracker that can be attached to any standard needle. The physician defines the target, desired needle path and skin entry point on a CT slice image. The accuracy of AngleNav was first tested in a 3D-printed calibration platform in a benchtop setting. An abdominal phantom study was then performed in a CT scanner to validate the accuracy of the device’s angular measurement. Finally, an in vivo swine study was performed to guide the needle towards liver targets (n = 8). CT scans of the targets were used to quantify the angular errors and needle tip-to-targeting distance errors between the planned needle path and the final needle position. The MEMS tracker showed a mean angular error of 0.01° with a standard deviation (SD) of 0.62° in the benchtop setting. The abdominal phantom test showed a mean angular error of 0.87° with an SD of 1.19° and a mean tip-to-target distance error of 4.89 mm with an SD of 1.57 mm. The animal experiment resulted in a mean angular error of 6.6° with an SD of 1.9° and a mean tip-to-target distance error of 8.7 mm with an SD of 3.1 mm. These results demonstrated the feasibility of AngleNav for CT-guided interventional workflow. The angular and distance errors were reduced by 64.4 and 54.8% respectively if using AngleNav instead of freehand insertion, with a limited number of operators. AngleNav assisted the physicians to deliver accurate needle insertion during CT-guided intervention. The device could potentially reduce the learning curve for physicians to perform CT-guided needle targeting.     

Using an inertial navigation algorithm and accelerometer to monitor chest compression depth during cardiopulmonary resuscitation

We present an original method using a low cost accelerometer and a Kalman-filter based algorithm to monitor cardiopulmonary resuscitation chest compressions (CC) depth. A three-axis accelerometer connected to a computer was used during CC. A Kalman filter was used to retrieve speed and position from acceleration data. We first tested the algorithm for its accuracy and stability on surrogate data. The device was implemented for CC performed on a manikin. Different accelerometer locations were tested. We used a classical inertial navigation algorithm to reconstruct CPR depth and frequency. The device was found accurate enough to monitor CPR depth and its stability was checked for half an hour without any drift. Average error on displacement was ±0.5 mm. We showed that depth measurement was dependent on the device location on the patient or the rescuer. The accuracy and stability of this small low-cost accelerometer coupled to a Kalman-filter based algorithm to reconstruct CC depth and frequency, was found well adapted and could be easily implemented.     

基于微电子机械加速度传感器的腹腔镜训练评价系统

我们设计了一种腹腔镜训练评价系统,并通过测试验证了系统的可行性。本研究选择了操作时间、平均速度、运动平稳度和空闲率作为评价系统的测量指标。利用MEMS加速度传感器系统完成运动参数采集。开发并设计MEMS加速度传感器模块,采取卡尔曼滤波算法对加速度计以及陀螺仪数据进行时域方面滤波,并通过欧拉角算法消除重力对加速度数据的影响;开发了指标计算软件平台,实现了由原始数据到指标参数的计算和数据保存。本研究通过对培训者10次培训操作过程的数据分析得到其操作时间训练前后减少了32%,左右手的操作平均速度分别降低45%和32%,左右手平稳度分别提高15%和19%,空闲率降低53%。     

Three-dimensional hall effect accelerometer for recording head movements of freely moving laboratory animals

A Hall effect device was constructed for a measurement of head movements in three spatial dimensions during classical conditioning experiments in cats. A Hall sensor was used to detect movements of a magnetic fragment floating in a small (15 x 15 mm) cube. The magnetic fragment was kept in the centre of the sealed cube with a thin coil spring which was filled with thin oil for damping excessive afteroscillations. A comparison of this device to a commercial accelerometer showed that the accuracy of the Hall device is sufficient for the movement recordings and that the device is sensitive also to slowly accelerating movements. The construction is compact and can be easily mounted, for example, on the head stage of a freely moving animal.     

Validity and reliability of triaxial accelerometers for inclinometry in posture analysis

There is a need for objective and quantitative methods for measuring posture and movement, so that, for instance, exposure-response relationships for work-related musculoskeletal disorders can be established. Inclinometry data have been obtained from triaxial accelerometers based on uniaxial solid-state accelerometers used in conjunction with a computer program to perform co-ordinate transformations. The transducer can be mounted in an arbitrary orientation on a body segment, since if two reference positions are recorded, the co-ordinate system of the transducer can be transformed to that of the body segment. The angular error of the system is small (1.3°), the reproducibility is high (0.2°), and the inherent angular noise is small (0.04°) and independent of the orientation of the device. Under quasi-static conditions, the angular velocities can be derived from the inclinometry data. The angular and the angular-velocity errors can be approximated using the relative deviation of the acceleration magnitude from gravitation. For applications involving a high degree of movement, the accelerometer data are still valid, although they cannot be interpreted as inclination. Used in combination with the computer program, the transducer can be used to measure posture and movement under static and quasi-static conditions, which occur in most areas of occupational work. It is shown that spherical co-ordinates can be used to present the inclinometry data.     

Support vector machine for classification of walking conditions using miniature kinematic sensors

A portable gait analysis and activity-monitoring system for the evaluation of activities of daily life could facilitate clinical and research studies. This current study developed a small sensor unit comprising an accelerometer and a gyroscope in order to detect shank and foot segment motion and orientation during different walking conditions. The kinematic data obtained in the pre-swing phase were used to classify five walking conditions: stair ascent, stair descent, level ground, upslope and downslope. The kinematic data consisted of anterior-posterior acceleration and angular velocity measured from the shank and foot segments. A machine learning technique known as support vector machine (SVM) was applied to classify the walking conditions. SVM was also compared with other machine learning methods such as artificial neural network (ANN), radial basis function network (RBF) and Bayesian belief network (BBN). The SVM technique was shown to have a higher performance in classification than the other three methods. The results using SVM showed that stair ascent and stair descent could be distinguished from each other and from the other walking conditions with 100% accuracy by using a single sensor unit attached to the shank segment. For classification results in the five walking conditions, performance improved from 78% using the kinematic signals from the shank sensor unit to 84% by adding signals from the foot sensor unit. The SVM technique with the portable kinematic sensor unit could automatically recognize the walking condition for quantitative analysis of the activity pattern.     

基于物联网的便携式肠内营养泵的设计

为实现医院肠内营养的统一管理和家庭肠内营养的便携式支持,本研究采用霍尔测速、光电检测、物联网技术,通过霍尔传感器、阵列式光耦传感器、ESP8266无线传感器,设计了一款便携式肠内营养泵,可根据用户设定的参数进行精准喂饲,且具有入口和出口堵塞及气泡检测报警功能。该营养泵具有喂饲精度高、检测灵敏度高、操作和管理方便、功耗低、便携等特点,非常适用于医院和家庭肠内营养支持,符合构建以物联网为基础的智慧医疗系统的趋势。     

Validation of an accelerometer for determination of muscle belly radial displacement

A commercial variable-capacitance micromachined accelerometer was validated for muscle belly radial displacement measurement. The displacement was calculated by the acceleration data being integrated twice and was compared with the results obtained simultaneously by an accurate mechanical displacement sensor based on an optical encoder. The aim of the investigation was to evaluate the accuracy and precision of an accelerometer for tensiomyography, which is a method for the detection of skeletal muscle contractile properties on the basis of muscle belly radial displacement. A hundred measurements at a bandwidth of 2300 Hz were performed. It was shown that the accuracy and precision in determination of the maximum displacement and the time of the maximum displacement from the calculated curve were satisfactory, in spite of the standard deviation of the twice-integrated acceleration growing approximately linearly with time. The results were accurate enough since the elapsed time from the beginning of the integration was small (less than 75 ms). The measured maximum displacement ranges were between 9.2 and 10.2 mm. The mean relative error was less than 1% (SD=0.02 mm) for the maximum displacement and about 1% (SD=0.6 ms) for the time to maximum displacement. The accuracy of the half-relaxation time determination was more uncertain because of the relatively high relative error of −2.4% (SD=3 ms). Results showed that a commercial micromachined accelerometer could be suitable for the measurement of muscle belly radial displacement and used for development of a future miniaturised and flexible system for the measurement of similar displacements.     

Monitoring Obstructive Sleep Apnea by means of a real-time mobile system based on the automatic extraction of sets of rules through Differential Evolution

Real-time Obstructive Sleep Apnea (OSA) episode detection and monitoring are important for society in terms of an improvement in the health of the general population and of a reduction in mortality and healthcare costs. Currently, to diagnose OSA patients undergo PolySomnoGraphy (PSG), a complicated and invasive test to be performed in a specialized center involving many sensors and wires. Accordingly, each patient is required to stay in the same position throughout the duration of one night, thus restricting their movements.This paper proposes an easy, cheap, and portable approach for the monitoring of patients with OSA, which collects single-channel ElectroCardioGram (ECG) data only. It is easy to perform from the patient’s point of view because only one wearable sensor is required, so the patient is not restricted to keeping the same position all night long, and the detection and monitoring can be carried out in any place through the use of a mobile device.Our approach is based on the automatic extraction, from a database containing information about the monitored patient, of explicit knowledge in the form of a set of IF…THEN rules containing typical parameters derived from Heart Rate Variability (HRV) analysis. The extraction is carried out off-line by means of a Differential Evolution algorithm. This set of rules can then be exploited in the real-time mobile monitoring system developed at our Laboratory: the ECG data is gathered by a wearable sensor and sent to a mobile device, where it is processed in real time. Subsequently, HRV-related parameters are computed from this data, and, if their values activate some of the rules describing the occurrence of OSA, an alarm is automatically produced.This approach has been tested on a well-known literature database of OSA patients. The numerical results show its effectiveness in terms of accuracy, sensitivity, and specificity, and the achieved sets of rules evidence the user-friendliness of the approach. Furthermore, the method is compared against other well known classifiers, and its discrimination ability is shown to be higher.